Cellular system optimized for mobile broadband such as, LTE and HSPA, uses link adaptation and HARQ. These functionalities are introduced for improved throughput performance both from a link and system perspective. In link adaptation, the terminals need to estimate the current radio performance and feedback that information to the network as a CQI. The network node (scheduler) then adapts the MCS. HARQ is then introduced for robust performance, and used for fast retransmission of erroneous decoded blocks. In a typical channel scenario, for optimised performance, a first transmission BLER around 10% is close to optimal from link perspective, and the scheduler is working to choose MCS based the CQI and BLER target.
Traditionally, CQI is solely based on SNR of the received signal. Internally generated decisions like adaptation/nulling/affecting of soft values to known interference are not taken into account, and thereby the CQI reported will not correctly describe the current decoding performance, and the reported CQI may be too optimistic. Thus, this will reduce the throughput in the system, by increasing the first transmission BLER, which typically should operate around 10% for optimized performance, and thereby increasing the packet retransmission rate. In some scheduling solutions some outer loop control is introduced in the scheduler that eventually will take such biased CQI into account, but such loop is normally slow and therefore the interference situation for the terminal could well be changed, for instance by a handover to another cell, before caution is taken by the controller. Hence optimized performance will not necessarily be reached using such outer loop BLER control compensation
Thus, the terminal estimates the CQI and reports it to the cellular system, and all reported CQIs are important for the cellular system for working efficiently.
WO 2005/000568 A2 discloses a method for biasing signal-to-interference ratio (SIR) to generate channel quality indicator (CQI) including measuring the packet error rate (PER) of a received signal and comparing the PER to the target PER to generate a correction term. The correction term is combined with the SIR estimation of a reference channel to generate a CQI. The CQI is reported to a transmitter to adjust signal configurations, such as code rate, modulation type, number of codes, power offset.
However, in modern cellular systems, numerous “tricks” are applied, both in the terminals and in the base stations, to improve performance. Some of these tricks may cause the estimated/reported CQIs to give an improper view of the actual channel qualities, which then may imply that the cellular system does not work that efficiently as intended. It is therefore a desire to provide a CQI that better reflects channel quality.